COS 97-3 - CANCELLED - Modeling with microbes: Improving predictions of how forest carbon and nitrogen cycles respond to declining nitrogen deposition

Abstract: As anthropogenic nitrogen (N) deposition declines across the eastern US, uncertainty remains in whether temperate forests will continue to provide a critically important carbon (C) sink. This uncertainty is amplified by the inability of ecosystem models to accurately capture the microbial mechanisms that drive greater soil C sequestration and N loss under enhanced N availability. Most conventional models do not explicitly represent soil microbes and as a result can’t capture ecosystem responses to N deposition that hinge on plant-microbe interactions, including weakening of priming effects that can increase soil C. Moreover, even microbially-explicit models that represent decomposers largely simplify N cycling by assuming that organic N follows C through stoichiometric controls and that inorganic N is transformed or lost with simple rate constant equations.



To address these limitations, we leveraged decades of C and N cycling data from a whole-watershed N fertilization experiment at the Fernow Experimental Forest in Parsons, WV to run the microbially-explicit plant-microbe interactions model FUN-CORPSE (Fixation and Uptake of Nitrogen- Carbon, Organisms, Rhizosphere, and Protection in the Soil Environment). We had three objectives, including: 1) Reproducing key ecosystem responses to N fertilization, 2) Modeling microbially-explicit inorganic N cycling, and 3) Assessing how modeled soil C and N retention respond to shifts in N deposition and other future climate drivers.



FUN-CORPSE accurately represents overarching ecosystem C and N cycling at the Fernow (e.g. soil C pools, streamwater N losses) and captures key responses to experimental N deposition, including the 25% decline in plant C cost of N acquisition that reduces root-primed decomposition and increases soil C. We validated our newly integrated inorganic N cycling in FUN-CORPSE against measured inorganic N pools, nitrification rates, and the relative abundance of nitrifying microbes. With microbially-explicit inorganic N cycling, FUN-CORPSE captures the 100% increase in nitrification rates and the 50% increase in streamwater nitrate loss under N fertilization. When we ran the model forward under projected declining N deposition, FUN-CORPSE predicted that N losses recovered more quickly than soil C pools. However, the eventual return of soil C to pre-fertilized levels indicates that additional C sequestered due to N deposition may be vulnerable to future loss. Collectively, this work demonstrates how modeling the microbes that drive soil C and N cycling informs model predictions of forest ecosystem responses to shifts in N deposition.